June 2005 — Exclusive
Print this article | Email this articleClick here to receive your FREE subscription to T.H.E. Journal
Electronic Full-Text Journal Articles: Convenience or Compromise
Keyword searching differs radically from subject heading searching, and yet few students recognize the difference. To assign subject headings, discipline experts read the articles and select descriptors from a standardized terminology. Once a user identifies the proper heading, he is rewarded with highly relevant citations. In contrast, the default keyword search means that every word in the title, in the abstract, in the subject headings, or even in the entire body of the article can retrieve the article. A search may retrieve thousands of resources, most of which are only marginally relevant to the topic. Rather than making research easier, this imprecise retrieval overwhelms the user. The resulting mishmash of unrelated articles is bewildering for the user and makes the sophisticated analysis of content unlikely.
Good Technology = Good Results
Most students are so comfortable with the Web that they believe their technological skills will automatically lead to successful searches-if they can create Weblogs and streaming videos, they can locate scholarly journal articles. Unfortunately, sophisticated computer skills do not translate into effective searching. Web-savvy researchers are much less likely to ask librarians for help with Web-based resources than they were back in the days of print indexes. Without assistance, students feel certain that they have located all available material. They do not realize that their results would vary dramatically if they changed their search terms or repeated their search in a more relevant database.
Full-Text That's Not Always Full
Because many aggregated databases provide the article text rather than page reproduction, graphical features disappear from the articles. And articles stripped of graphs, photographs, charts, special characters, tables, or illustrations are not the same as the original versions. Even more deplorable, full-text databases do not always include every article from each issue of a journal. Cover-to-cover full-text is rarer than most vendors lead their customers to believe. Letters to the editor, book reviews, editorials, p'etry, and brief articles are routinely skipped, while in some instances, so are major articles. When the Supreme Court ruled in the case of New York Times Co. Inc. v. Tasini that freelance authors were entitled to additional payment when their work was republished electronically, publishers responded by pulling thousands of articles from online databases. Interestingly, comparisons of printed journals and their articles' availability in full-text databases have shown an inclusion rate as low as 60 percent.
Currency
Users are often misled about the currency of the material offered in electronic databases. Although vendors load articles into their databases on a daily basis, print and electronic versions of the same journals are not necessarily issued simultaneously. Indexing sometimes appears weeks before the full article. Many journals also embargo electronic access to their recent issues as an incentive to retain their print subscribers. It is often difficult to identify these embarg'ed or tardy journals from the lists provided by vendors.
Recommendations
The delivery of full-text documents to desktops is a phenomenon that is not only popular, but one that will alter the world of research over the next decade. This information revolution holds the potential to empower both the sophisticated and the inexperienced. No serious educator or student would ever want to retreat back a decade or two into the "dark ages" of information technology.
